1/*
2 * Altivec optimized MP3 decoding functions
3 * Copyright (c) 2010 Vitor Sessak
4 *
5 * This file is part of FFmpeg.
6 *
7 * FFmpeg is free software; you can redistribute it and/or
8 * modify it under the terms of the GNU Lesser General Public
9 * License as published by the Free Software Foundation; either
10 * version 2.1 of the License, or (at your option) any later version.
11 *
12 * FFmpeg is distributed in the hope that it will be useful,
13 * but WITHOUT ANY WARRANTY; without even the implied warranty of
14 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
15 * Lesser General Public License for more details.
16 *
17 * You should have received a copy of the GNU Lesser General Public
18 * License along with FFmpeg; if not, write to the Free Software
19 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
20 */
21
22#include "config.h"
23#include "libavutil/attributes.h"
24#include "libavutil/cpu.h"
25#include "libavutil/internal.h"
26#include "libavutil/mem_internal.h"
27#include "libavutil/ppc/cpu.h"
28#include "libavutil/ppc/util_altivec.h"
29#include "libavcodec/mpegaudiodsp.h"
30
31#if HAVE_ALTIVEC
32
33#define MACS(rt, ra, rb) rt+=(ra)*(rb)
34#define MLSS(rt, ra, rb) rt-=(ra)*(rb)
35
36#define SUM8(op, sum, w, p)               \
37{                                         \
38    op(sum, (w)[0 * 64], (p)[0 * 64]);    \
39    op(sum, (w)[1 * 64], (p)[1 * 64]);    \
40    op(sum, (w)[2 * 64], (p)[2 * 64]);    \
41    op(sum, (w)[3 * 64], (p)[3 * 64]);    \
42    op(sum, (w)[4 * 64], (p)[4 * 64]);    \
43    op(sum, (w)[5 * 64], (p)[5 * 64]);    \
44    op(sum, (w)[6 * 64], (p)[6 * 64]);    \
45    op(sum, (w)[7 * 64], (p)[7 * 64]);    \
46}
47
48static void apply_window(const float *buf, const float *win1,
49                         const float *win2, float *sum1, float *sum2, int len)
50{
51    const vector float *win1a = (const vector float *) win1;
52    const vector float *win2a = (const vector float *) win2;
53    const vector float *bufa  = (const vector float *) buf;
54    vector float *sum1a = (vector float *) sum1;
55    vector float *sum2a = (vector float *) sum2;
56    vector float av_uninit(v0), av_uninit(v4);
57    vector float v1, v2, v3;
58
59    len = len >> 2;
60
61#define MULT(a, b)                         \
62    {                                      \
63        v1 = vec_ld(a, win1a);             \
64        v2 = vec_ld(b, win2a);             \
65        v3 = vec_ld(a, bufa);              \
66        v0 = vec_madd(v3, v1, v0);         \
67        v4 = vec_madd(v2, v3, v4);         \
68    }
69
70    while (len--) {
71        v0 = vec_xor(v0, v0);
72        v4 = vec_xor(v4, v4);
73
74        MULT(   0,   0);
75        MULT( 256,  64);
76        MULT( 512, 128);
77        MULT( 768, 192);
78        MULT(1024, 256);
79        MULT(1280, 320);
80        MULT(1536, 384);
81        MULT(1792, 448);
82
83        vec_st(v0, 0, sum1a);
84        vec_st(v4, 0, sum2a);
85        sum1a++;
86        sum2a++;
87        win1a++;
88        win2a++;
89        bufa++;
90    }
91}
92
93static void apply_window_mp3(float *in, float *win, int *unused, float *out,
94                             ptrdiff_t incr)
95{
96    LOCAL_ALIGNED_16(float, suma, [17]);
97    LOCAL_ALIGNED_16(float, sumb, [17]);
98    LOCAL_ALIGNED_16(float, sumc, [17]);
99    LOCAL_ALIGNED_16(float, sumd, [17]);
100
101    float sum;
102    int j;
103    float *out2 = out + 32 * incr;
104
105    /* copy to avoid wrap */
106    memcpy(in + 512, in, 32 * sizeof(*in));
107
108    apply_window(in + 16, win     , win + 512, suma, sumc, 16);
109    apply_window(in + 32, win + 48, win + 640, sumb, sumd, 16);
110
111    SUM8(MLSS, suma[0], win + 32, in + 48);
112
113    sumc[ 0] = 0;
114    sumb[16] = 0;
115    sumd[16] = 0;
116
117    out[0  ]  = suma[   0];
118    out += incr;
119    out2 -= incr;
120    for(j=1;j<16;j++) {
121        *out  =  suma[   j] - sumd[16-j];
122        *out2 = -sumb[16-j] - sumc[   j];
123        out  += incr;
124        out2 -= incr;
125    }
126
127    sum = 0;
128    SUM8(MLSS, sum, win + 16 + 32, in + 32);
129    *out = sum;
130}
131
132#endif /* HAVE_ALTIVEC */
133
134av_cold void ff_mpadsp_init_ppc(MPADSPContext *s)
135{
136#if HAVE_ALTIVEC
137    if (!PPC_ALTIVEC(av_get_cpu_flags()))
138        return;
139
140    s->apply_window_float = apply_window_mp3;
141#endif /* HAVE_ALTIVEC */
142}
143